Brave New Stepfamilies

01-Stewart.qxd 11/28/2006 6:29 PM Page 1

hat comes to mind when you hear the word stepfamily? Before you Wanswer, consider the following scenarios:

• Rhonda and her boyfriend, Al, have a child, Emily. The couple’s relationship doesn’t work out and Al moves out of state. He has no contact with his daughter and pays no child support. Rhonda meets Peter and they marry. Peter adopts Emily. • Carol and Randy are divorced. Carol has custody of their two children, Emma and Sophia. After a few months of dating, Carol’s boyfriend, Roger, moves in with Carol and the girls. • Bobby lives with his mother, Elaine. Bobby’s father, Doug, has remarried Leslie. The couple lives with Leslie’s children from her first marriage, Teddy, Austin, and Abbey. Bobby sees his father every other weekend, usually for a movie and a bite to eat. Sometimes Leslie and her children go along (depending on what’s showing). • Janet is married to Ron and the couple has two sons, Billy and Justin. Janet falls in love with Ann, a coworker, and divorces Ron. Ann moves in with Janet, Billy, and Justin. The boys refer to her as “Aunt Ann.” • Rosemary is divorced and has three college-aged children, Sarah, Ben, and Lily. Rosemary meets James, a widower, at a retirement party and after a two-week courtship, they fly off to Las Vegas and get married. James has a grown son, Todd. The kids get together and plan a recep- tion for the couple.

1 01-Stewart.qxd 11/28/2006 6:29 PM Page 2

2 BRAVE NEW STEPFAMILIES

Which of these families is a stepfamily? Which family members did you include, and what criteria did you use to decide? The opinions of undergradu- ate students on this issue vary widely. The discussion typically goes like this: “Definitely Rhonda and Peter, because he adopted Emily, but not Rosemary and James, because the kids are grown up and out of the household.” Or, “‘Yes,’ to Bobby, because he sees his father pretty frequently, but ‘no’ to Carol and Roger, because they are not married.” Then, someone inevitably pipes up, “But using marriage as a criterion is not fair to Janet and Ann, because, being lesbians, they can’t legally marry (in most states).” Students tend to enjoy this debate, because it gets at the core of our very personal and deeply held ideas about the meaning of family in American society. Many students offer the sto- ries of friends and relatives who are in the same situations as the people described in the vignettes. Many students talk about being in, or having been in, a similar situation themselves. I have never had a class be in complete agreement about any of them. You may be surprised to find out that, among family scholars, there is also a distinct lack of agreement about what constitutes a stepfamily. Moreover, an examination of published research on stepfamilies would lead one to conclude that the majority of stepfamily scholars do not consider any of these situations stepfamilies. Despite variation in legal marriage, residential status, sexual orientation, and the rest, these different scenarios have something important in common: the absence of a biological relationship between one’s children and one’s serious romantic partner. In this book, I argue that this is what makes a stepfamily.1 What social and demographic shifts are causing this to happen? Among them are a growth in the number of women having children outside of marriage, an increase in couples with children “living together,” increases in nonresident parent involvement and the shared physical custody of children after divorce, the aging of the population, increasing racial and ethnic diversity, and increasing awareness and tolerance of lesbian and gay relationships. I propose that family scientists (and the public) adopt a broader definition of the stepfamily, one that acknowledges these social and demographic shifts, highlights stepfamily diversity, and reflects the current realities of Americans’ lives. In fact, under a broader definition of the stepfamily, remarried couples with stepchildren under 18 living in the household (the traditional definition of a stepfamily) would actually account for a minority of all stepfamilies (Stewart, 2001). 01-Stewart.qxd 11/28/2006 6:29 PM Page 3

Brave New Stepfamilies 3

Photo 1.1 Stepfamilies of the 21st century face many new challenges. © LWA-Dann Tardif/Corbis.

I am not the first to take a broader perspective on stepfamilies. In their most recent book, leading stepfamily scholars Lawrence Ganong and Marilyn Coleman make a similar argument (Ganong & Coleman, 2004). Family demographer Larry Bumpass and his colleagues did the same a decade ago when they proposed that stepfamilies incorporate trends in out-of-wedlock childbearing and cohabitation (Bumpass, Raley, & Sweet, 1995). It is important to note that while many stepfamily researchers may have wanted to use a broader definition 01-Stewart.qxd 11/28/2006 6:29 PM Page 4

4 BRAVE NEW STEPFAMILIES

of stepfamilies, they were constrained by the definition used by our official source of information on the United States population, the U.S. Census. Tradi- tionally, the U.S. Census has defined stepfamilies in terms of marriages and households, and this practice continues today (U.S. Census, 2005a, 2005b). The purpose of this book is to examine new and diverse kinds of stepfamilies and to discuss what is known, and what is unknown, about them in relation to our current knowledge base. Students and researchers will receive basic information about stepfamily structure, theory, dynamics, and processes while being exposed to “invisible” stepfamilies and “hidden” stepfamily members. This book is unique in its breadth, macrolevel perspective, and attention to diversity in the areas of family structure, gender, age, race and ethnicity, and sexual identity.

THE EVOLUTION OF AN INSTITUTION

Social, economic, and demographic changes have occurred over the 20th century, especially in the past three decades or so, which have produced funda- mental changes in the nature stepfamily life. However, it is difficult to think about the evolution of the American stepfamily without first considering how the American family itself has evolved and is evolving. Like the stepfamily, the American family has become more diverse and more complex. Consequently, we now realize that there is no, and has never been, a “standard” American family (Coontz, 1992). The term family more than ever before represents a mosaic of family forms. In fact, the so-called traditional family—the kind characterized so humor- ously by The Simpsons, with the female parent focused on the home and domes- tic tasks and the male parent focused on earning an income—is now a distinct minority (Kain, 1990). According to data from the 2002 Current Population Survey (CPS), traditional breadwinner-homemaker families like these repre- sented only 7% of all U.S. households (Population Reference Bureau, 2003; see Figure 1.1). Today, no one family form dominates over all others. One often hears of the “divorce revolution” as the primary reason for this situation, but trends in divorce and other family-related patterns are themselves the effects of still other social forces. Drawing from a wide range of longitudinal data, Andrew Cherlin (1992) pieced together an explanation for the American family’s long-term trend toward diversification. Cherlin argued that technological advancement in the United States (industrialization followed by growth in 01-Stewart.qxd 11/28/2006 6:29 PM Page 5

Brave New Stepfamilies 5

Traditional 7% Dual-income, with children 16%

Dual-income, no children 13%

Other 64%

Figure 1.1 Types of U.S. Households, 2002

SOURCE: Ameristat analysis of data from the 2002 Current Population Survey (March supplement) (Population Reference Bureau, 2003). Used with permission of Population Reference Bureau. Copyright © 2003. U.S. Nuclear Meltdown. Top Panel only, “Types of U.S. Households, 2002.” NOTE: “Traditional” households include married-couple households with children in which only the husband is in the labor force.

service-type jobs) has gradually changed women’s work roles, increasingly imbuing them with economic opportunities that give married women greater economic independence from their husbands and an incentive for leaving an unhappy marriage, thereby promoting a long-term rise in the divorce rate. It is difficult for us, in the first decade of a new millennium, to appreciate the significance of the technological change behind this trend. Most of us have no concrete impression of living in a “household of production,” in which all family members, including children, contribute to their own survival by work- ing. It is hard to relate to a time when fathers and sons created food through farming and animal domestication, and mothers and daughters turned slaugh- tered hogs into sausages and spun wool into yarn. Commercial enterprises take care of those tasks for us, and work usually takes place outside the home. Consequently, with our cookouts and holiday celebrations, DVDs, digital television, cell phones, and online shopping, we center our families on consump- tion rather than on production, and husbands’ and wives’ economic roles are no longer as interdependent as they once were (Hutter, 1998). Marriage is rarely necessary for economic survival, and divorce less often results in economic devastation. Consequently, men and women are no longer as compelled to stay married or get married in the first place. 01-Stewart.qxd 11/28/2006 6:29 PM Page 6

6 BRAVE NEW STEPFAMILIES

Trends in Divorce

Figure 1.2 shows the historical trend in the divorce rate over the last century or so as presented by Cherlin (1992). The long-term rise in the divorce rate has exhibited changes that suggest that more than technological change has driven family evolution. Cherlin showed that from 1860 to the end of World War II, the divorce rate rose steadily, aside from three brief fluctuations: a tiny spike just after the Civil War, another somewhat larger spike after World War I, and a temporary lowering of divorce during the Great Depression. After WWII, however, the annual divorce rate exhibited three longer, wider swings: a dra- matic increase starting a few years after WWII, a decline during the “good old days” of the 1950s, and a steady rise from the 1960s to the end of the 1970s.

24 22 20 18 16 14 12 10 8 Annual divorce rate 6 4 2 0 1860 1880 1900 1920 1940 1960 1980 Year

Figure 1.2 U.S. Annual Divorce Rate, 1860–1990

SOURCE: From Cherlin, A., Marriage, Divorce, Remarriage, copyright © 1992. Reprinted by permission of Harvard University Press. Original data from Paul H. Jacobson (1959), American Marriage and Divorce (New York: Reinhart), Table 42; National Center for Health Statistics ([NCHS] 1973), 100 Years of Marriage and Divorce Statistics, Table 4; NCHS, “Advance Report of Final Divorce Statistics, 1987,” Monthly Vital Statistics Report, 38(12), Table 1; NCHS, “Annual Summary of Births, Marriages, Divorces, and Deaths: United States, 1989,” Monthly Vital Statistics Report, 38(13). NOTE: For 1860–1920, divorces per 1,000 existing marriages; for 1920–1988, divorces per 1,000 married women 15 years old and older. 01-Stewart.qxd 11/28/2006 6:29 PM Page 7

Brave New Stepfamilies 7

Cherlin noted that historical events that affect the attitudes of a cohort of people as they are growing up, as well as accelerations and decelerations in the evolution of other social phenomenon, could slow, speed up, or reverse trends in family evolution. For instance, drawing heavily from the work of sociologist Glen Elder, Jr., and economist Richard Easterlin, Cherlin argued that during the Great Depression, the family became something of a scarce resource (Cherlin, 1992; Easterlin, 1978, 1980; Elder, 1974). Many young men and women delayed getting married and having children during these hard times because it was too expensive for them to set up housekeeping. For instance, some young boys became their family’s main breadwinner when their fathers lost their jobs. Once the economy turned around after WWII, people who were deprived of, or saw others deprived of, a family of their own during the Depression were finally able to act on this value. Consequently, the divorce rate reversed its trend, drop- ping below what the prior long-term trend would have predicted for the 1950s. Trends in the birth rate and age at marriage reversed as well, as people began to marry sooner and have children earlier. Social programs such as the GI Bill, which allowed veterans to go to college and buy homes, and a culture of domes- ticity (television shows and magazines celebrating the hearth, home, and traditional gender roles) encouraged marriage and family formation even more. Cherlin speculated that the divorce rate reversed again in the 1960s and 1970s because of a sharp increase in the rate at which women were participat- ing in the labor force and a rise in women’s wages. Thus, although the steady rise in the proportion of women in the labor force had long been shaping American family patterns, its historically unusual jump in the 1960s and 1970s made it even more influential in that period. Suddenly many more women had a means of surviving a divorce and, it is presumed, decided to end their marriages (Ruggles, 1997). During the later part of this period, a visible feminist movement, the second in American history, arose to not only fight discrimination against women, but to also encourage women to pursue their educations and professional careers (Collins & Coltrane, 1991). It is important to note that the data presented in Figure 1.2 has not been extended beyond 1988. The National Center for Health Statistics (NCHS), the main source of data on marriage and divorce, stopped collecting and reporting detailed marriage and divorce statistics in the early 1990s due to budgetary pressures and limitations in the information collected by some states (NCHS, 2005). However, a more recent analysis based on vital statistics data from the NCHS and the CPS indicate that the divorce rate has remained more or less stable since the late 1980s (Goldstein, 1999). 01-Stewart.qxd 11/28/2006 6:29 PM Page 8

8 BRAVE NEW STEPFAMILIES

Deciphering just how much the historical rise in divorce was due to women’s greater economic independence, the second wave of feminism, or the declining labor-market opportunities of men is a subject of debate (Oppenheimer, 1997). Nevertheless, as evidenced by a sharp increase from the late 1960s through the mid-1970s (Figure 1.3) in the proportion of Americans who said that they would vote for a woman for president, one indicator of women’s independence presented by Cherlin (1992), women during this period do seem to have gained a much wider sense of freedom from traditional roles.

100 90 80 70 60 Percent 50 40 30 20 1937 1949 1958 1971 1974 1977 1982 1985 1988 1990 1993 1996 1998 Year

Men Women

Figure 1.3 Percentage of Americans Who Would Vote for a Woman for President

SOURCE: Adapted from Cherlin, A., Marriage, Divorce, Remarriage, copyright © 1992. Reprinted with permission of Harvard University Press. Original data from George H. Gallup (1972), The Gallup Poll: Public Opinion 1935–1971, Vols. 1–3 (New York: Random House); GSS Cumulative Datafile 1972–2004, Full Analysis, Survey Documentation and Analysis Archive, University of California, Berkeley, 2005.

There is a lack of empirical support for the idea that increases in women’s economic prospects (education, employment, earnings) are disruptive to marriages among more recent cohorts of men and women (post-1968). For instance, some dual-earner couples have more enduring marriages than male-breadwinner couples (Ono, 1998). Indeed, research indicates that, among more recent cohorts of Americans, women’s economic contributions are positively, rather than nega- tively, related to marriage (Sweeney, 2002). Nock (1998b) suggests that the 01-Stewart.qxd 11/28/2006 6:29 PM Page 9

Brave New Stepfamilies 9

erosion of male earnings and gains in women’s earnings may mean that American couples are once again, as they were in preindustrial America, economically interdependent. Other factors that may have contributed to increases in divorce are the birth control pill, which allowed women to control their reproduction and decrease their dependency on marriage; no-fault divorce laws, which makes divorce easier to obtain and reduces stigma (Friedberg, 1998); and increased longevity. Interestingly, the average length of a marriage hasn’t actually changed that much over the century; the difference is that now marriages are broken by divorce rather than death (Davis, 1972). One hypothesis is that, faced with the prospect of remaining in an unhappy marriage for longer and longer periods, couples may have increasingly chosen to end their marriages before “death do us part.” Scholars also point out that Americans’ increasing geographic mobility has weakened the influence of a variety of institutions, especially religion, over the family (Hutter, 1998). As mentioned above, the divorce rate has stabilized at a high level. It is estimated that between four and five of every 10 marriages formed in the 1990s will end in divorce (Norton & Miller, 1992). Because most people who divorce remarry (Furstenberg & Cherlin, 1991), and most remarriages involve children from a prior marriage (Cherlin & McCarthy, 1985), the rise in divorce plays a major role in explaining the growth in nontraditional families including stepfamilies. People marrying, having children, divorcing, cohabiting, marrying again, and perhaps again, results in complicated and constantly changing kinship structures. But remarriage is nothing new. Subject to the influence of local laws, reli- gious customs, and the availability of potential spouses, remarriage after the death of a spouse has been common throughout American history (Ihinger- Tallman & Pasley, 1987). Data for earlier periods of American history are sketchy. Analyzing 1689 census data from the Plymouth Colony town of Bristol, John Demos (1966) noted, “Unfortunately, little is known about the origin and purpose of the [1689] census—by whom it was compiled, and for what purpose” (p. 64). Recognizing the limitations of his data, he estimated that, of Bristol’s population of people at least 50 years old, 40% of the men and 26% of women remarried at least once. These remarriages took place quickly. Customarily, a widower or widow remarried within a year of a spouse’s death. Data from the English village of Clayworth (Nottinghamshire) in 1688 indicate that roughly one in six households was a stepfamily (Phillips, 01-Stewart.qxd 11/28/2006 6:29 PM Page 10

10 BRAVE NEW STEPFAMILIES

1997). What is different now is that the long-term rise in the divorce rate has caused the American stepfamily to evolve from a family form commonly produced by the death of a spouse to a family form commonly produced by divorce. Our most recent estimate, based on the 1980 and 1985 CPS, indicates that 90% of stepfamilies are formed through divorce (Bumpass, Sweet, & Castro Martin, 1990).

Social and Demographic Changes in Family Life

Having established that premise, I move on to the focus of this book. Namely, the social and demographic shifts that call for another change in perspective. The stepfamily is no longer merely the product of divorce or the death of a spouse. Below, I review social and demographic trends that have affected the structure and composition of stepfamilies.

Nonmarital Childbearing

The first trend to affect the composition of the family is unmarried childbearing, or women having children outside of marriage. Formerly known as “illegitimacy,” this family pattern has lost much of its stigma and has become widespread. Related to women and men delaying getting married and married women having fewer children, statistics indicate that one third of all births today are to unmarried women (Wu & Wolfe, 2001). These patterns vary dra- matically by race and ethnicity: 22% for Whites, 42% for Hispanics, 69% for African Americans, 59% for American Indians, and 16% for Asians/Pacific Islanders (Bianchi & Casper, 2000). What do these figures mean for stepfamilies? Because almost all women (about 90%) marry eventually (Goldstein & Kenney, 2001), a large proportion of stepfamilies are formed through first marriages rather than remarriages. Even as far back as the early 1980s, one third of children entering a stepfamily did so after being born to an unmarried mother rather than through parental divorce (Bumpass et al., 1995). Childbearing outside of marriage can occur to previously married women too. An analysis of births occurring between 1970 and 1984 to women who were part of the National Survey of Families and Households indicates that 40% of nonmarital births were to women who had been previously married (Bumpass & Sweet, 1989a). Among U.S. women who had their first child in 2001, 7% were to currently divorced or widowed women (Downs, 2003). I refer to stepfamilies in which children 01-Stewart.qxd 11/28/2006 6:29 PM Page 11

Brave New Stepfamilies 11

from previous relationships were born outside of marriage as stepfamilies created by nonmarital childbearing.

Cohabitation

It is also clear that stepfamily relationships cannot be identified through marriage alone. Cohabitation has transformed American families, including stepfamilies. You have no doubt observed that “living together” has become extremely common. Related to the same broad economic and cultural changes that underlie nonmarital childbearing (e.g., expansion of women’s work roles, sexual freedom, increasing individualism and secularization), the majority of young men and women will cohabit at some point in their lives (Smock, 2000). Estimates of how cohabitation has affected stepfamily living have not been updated and are based on data collected in the 1980s. Those estimates suggest that over half (56%) of all couples marrying between 1990 and 1994 cohabited before marrying (Bumpass & Lu, 2000). This is not just a phenomenon among young, childless couples. Cohabitation is even more common before remarriage than before first marriage. In the 1980s, roughly two thirds of remarriages were preceded by cohabitation compared to one third in the 1970s (Bumpass & Sweet, 1989b). Moreover, 40% of cohabitations involve children (Fields & Casper, 2001; Simmons & O’Connell, 2003). Some of these are children born to cohabiting couples and some are children from previous unions. In the late 1980s, one quarter of cohabiting couples had children from previous relationships living in the household (Bumpass, Sweet, & Cherlin, 1991). If living situations like these were considered to be a kind of stepfamily, they would account for one quarter of all stepfamilies (Bumpass et al., 1995). Given the growth in cohabitation since the 1980s (Bumpass & Lu, 2000), the proportion of stepfamilies that are cohabiting is probably even greater today. Stepfamilies formed through cohabitation rather than marriage are referred to as cohabiting stepfamilies.

Residential Living Arrangements

Still other social forces are acting upon stepfamilies. Take something as basic as where the members of a family live. We tend think of families and households as synonymous, but for stepfamilies, residence is dynamic, with people continually shifting from one household to another (Coleman, Ganong, & Fine, 2000). Seeing stepfamilies as extending beyond a single household has 01-Stewart.qxd 11/28/2006 6:29 PM Page 12

12 BRAVE NEW STEPFAMILIES

become important because, increasingly, both parents are staying involved with the children after divorce, rather than just one, usually the children’s the mother. If we consider adults with part-time or “visiting” stepchildren to be stepparents, they would make up over half of all stepparents nationally (Stewart, 2001). Shifting our focus to include nonresident family members would also necessitate more studies of stepmothers. Because the majority of children reside with their biological mothers after divorce, most stepmothers do not live with their stepchildren full time. Viewing stepfamilies as spanning households challenges traditional conceptualizations of the family, common notions about what constitutes parenthood, explanations of a variety of family phenomena, and many of our social policies that limit a family to people who live together. Chapter 6 therefore examines multihousehold stepfamilies.

Increasing Racial and Ethnic Diversity Increasing racial and ethnic diversity in American society, and increasing awareness of and attention to this diversity, is also influencing our under- standing of stepfamily life. Lack of attention to race and ethnicity in the past caused family scholars to reach erroneous conclusions about the historical prevalence of African American stepfamilies and the forces underlying their formation (for an example of this type of error, see Coleman’s [1994] criticism of Popenoe). Indeed, although the historical death-to-divorce transition in the stepfamily formation process applies to a broad base of the American population, divorce has long been a common precursor to stepfamily formation among spe- cific racial and ethnic minorities. Moreover, a consideration of nonmarital childbearing, cohabitation, and nonresident parenthood has important implications for racial and ethnic groups who have disproportionately high rates of each. Yet the bulk of family and stepfamily research has been conducted using White families or does not adequately account for racial differences. Since Census 2000, we have had an especially increased awareness that American society has become much more diverse than simply “black and white” (McLoyd, Cauce, Takeuchi, & Wilson, 2000, p. 1070). Although other minorities such as Hispanics and Asians will make up an increasing proportion of the U.S. population, including the stepfamily population, there is not enough research on these stepfamilies for solid reviews. For the time being, the only kind of racial and ethnic diversity in stepfamilies covered in this book is African American stepfamilies (Chapter 7). 01-Stewart.qxd 11/28/2006 6:29 PM Page 13

Brave New Stepfamilies 13

Increased Awareness and Tolerance of Gays and Lesbians

Stepfamilies with gay or lesbian parents are formed when a man or a woman with children from a previous heterosexual (or homosexual) relationship forms a romantic union with a partner of the same sex. The new partner of the children’s biological or adoptive parent becomes the children’s stepparent in the same way that heterosexual partners do. In the case of gays and lesbians, it is important to be clear that to be a stepfamily, the child(ren) must be from a previous relationship of one of the partners. In other words, I do not consider gays and lesbians who together adopt or bear a child (who is biolog- ically or legally related to just one partner) a stepfamily because that child is a product of the current relationship. Stepfamilies with gay and lesbian parents warrant our attention, especially as the public’s attitudes toward them change their opportunities to create, or in some cases publicly acknowledge, family relationships. Such attention will help us explain family dynamics that sociologists have overlooked and that may become more prevalent in the future. While gays and lesbians still face system- atic violence and discrimination (Waldo, Hesson-McInnis, & D’Augelli, 1998), Americans’ opposition to adoption by gays and lesbians has weakened. For instance, the percentage of Americans who say that homosexuality is “always wrong” has declined from 75% in 1987 to 56% in 1998 (Loftus, 2001); nearly half of Americans (48%) report having close personal friends or relatives who are gay or lesbian (Taylor, 2000); less than a third of Americans think that gays cannot parent as well as straight people can (Price, 1998); and two thirds of American 2001 high school graduates favor legal recognition of same-sex marriages (Gilbert et al., 2001).

Aging of the Population

Chapter 9 investigates stepfamilies with adult stepchildren, defined as stepfamilies that have children over 18 years old. There are two modes of entry into this kind of stepfamily: (1) the “aging” of stepfamilies formed when the children were young, so that the stepchildren are now adults, and (2) the parents of adult children forming unions with new spouses and partners. The American population is aging, meaning that a growing percentage of Americans are reaching retirement age and becoming elderly. What is signifi- cant about this is that for the first time in American history, substantial numbers of older Americans have experienced divorce, remarriage, and stepfamily life. 01-Stewart.qxd 11/28/2006 6:29 PM Page 14

14 BRAVE NEW STEPFAMILIES

Rates of divorce, nonmarital childbearing, and stepfamily living were high for the massive Baby Boom generation (the roughly 80 million people who were born between 1946 and 1964), which reached adulthood and formed families in the 1970s and 1980s. The stepfamilies formed during this period are aging along with the population, and an unprecedented number of older Americans are stepparents (Cornman & Kingson, 1996). Moreover, the children and stepchildren of the “divorce generation” have, by now, become adults themselves and are in the process of forming (and sometimes, dissolving) their own families. Thus it has become imperative that we track changes in stepfamily relationships and processes over the lifecourse, paying special attention to stepfamily relationships among adult stepchildren, their elderly parents and stepparents, and stepgrandparents and stepgrandchildren. Concomitantly, the divorce and remarriage rates among older adults have increased (Cooney, 1993; Holden & Kuo, 1996) and cohabitation among older Americans is becoming increasingly common as well (Chevan, 1996; King & Scott, 2005). How remarriage and stepfamily formation affects parents’ and stepparents’ relationships with adult children and grandchildren therefore requires some consideration. Chapter 10 focuses on the implications of the emergence of these new stepfamily forms. I examine how well they mesh with current research and theory, the practices of therapists, school administrators, and other profession- als who work with stepfamilies, and family policy. Acknowledging issues such as parenting across households, nonmarital relationships, and racial and ethnic diversity means that seemingly straightforward questions such as, “Should a stepfather discipline his stepchild?” will have complex answers.

A New Model of Stepfamilies

The social and demographic trends described above have important implications for the way that stepfamilies are defined. Table 1.1 compares the traditional definition of a stepfamily to a “revised” definition that incorporates the above trends. Whereas the focus of most previous studies of stepfamilies has been on remarriage, the new model also includes first-married and cohabiting couples with stepchildren. Whereas the traditional definition emphasizes stably situated coresident stepchildren, the new model incorporates nonresident stepchildren living in other households and shifts in residence over time. Whereas the traditional definition focuses on parenting young, school-aged, and adolescent children, the revised model expands parenting to adult children and examines parent-child relationships across the lifecourse. Finally, whereas 01-Stewart.qxd 11/28/2006 6:29 PM Page 15

Brave New Stepfamilies 15

Table 1.1 Traditional Versus Revised Definition of Stepfamilies

Assumption Traditional Revised

Union type Remarriage First marriage, remarriage, cohabitation

Residence of children Coresident, static Coresident and nonresident, dynamic

Stage of family Childrearing, children Parenting across the life cycle 0–18 years lifecourse (includes children 18+ years)

Race and ethnicity White White, African American, Hispanic, etc.

Social class Middle class All classes (lower, middle, upper)

Sexual orientation Heterosexual (“straight”) Heterosexual or homosexual (gay or lesbian)

the majority of studies focus on White, middle-class, heterosexual stepfamilies, this new model emphasizes racial, ethnic, and socioeconomic diversity and both heterosexual and same-sex couples. There are many more studies of traditional stepfamilies than nontraditional stepfamilies. In preparation for this book, I reviewed 271 journal articles, books, and book chapters on stepfamilies (only those that provide details concerning the study’s sample have been included) published in the last three decades. Of these, 230 studies (85%) defined stepfamilies in the traditional way. In contrast, only 41 studies (15%) contained information on newer stepfamily forms (analysis available upon request).

THE PROBLEM OF ESTIMATING THE PREVALENCE OF STEPFAMILIES

Estimating the pervasiveness of stepfamily living in the United States should be straightforward. Yet, as is the case with stepfamily research in general, deter- mining how many stepfamilies there are turns out to be quite complicated. First, arriving at an estimate of stepfamilies depends on how the researcher chooses to 01-Stewart.qxd 11/28/2006 6:29 PM Page 16

16 BRAVE NEW STEPFAMILIES

define them, and these definitions vary between data sources and studies. Second, there have been surprisingly few “official” estimates of stepfamilies. Those that exist have been pieced together from data from the U.S. Census, the NCHS, and various national surveys of American families including the CPS, National Survey of Families and Households (NSFH), Survey of Income and Program Participation (SIPP), and the National Survey of Family Growth (NSFG; Bramlett & Mosher, 2002). Relying on and sometimes combining different data sources results in not a single estimate of stepfamilies, but a number of statistics whose usefulness will depend on the particular point the researcher is trying to make. A third problem is that, due to budget limitations, government agencies (e.g., the U.S. Census and NCHS) are no longer producing the detailed reports of trends in marriage, divorce, and stepfamilies. The marriage and divorce reg- istration system of the NCHS was discontinued in the 1990s (NCHS, 2005), making the detailed marital history information of the SIPP and NSFG our leading sources of information on the prevalence of stepfamilies (Kreider, 2005). Because these data take time to clean and analyze, there is a substantial lag between the completion of the survey and release of the information to the public. Although Cycle 6 of the NSFG was collected in 2002, the most recent comprehensive report of marriage and family patterns based on this survey come from Cycle 5, collected in 1995. The result is that many of our most commonly cited statistics on stepfamilies may seem outdated, having been based on data collected in the late 1980s and early 1990s. These estimates have not been revised using newer data, and it is unclear how accurate they are for describing stepfamily life today. Given the dynamic and complex nature of American family life today, stepfamilies are in general difficult to capture. Therefore, any statistics on stepfamilies are going to be conservative, under- estimating stepfamilies rather than overestimating them.

Stepfamilies Created by Divorce and Remarriage

Let us begin with some statistics that reflect the most restrictive definition of stepfamilies: remarried couples who have children (of either spouse or both) from previous marriages living in the household. The children must be under age 18. There were 3.2 million marriages in the United States in 2001 (NCHS, 2004). According to the 2001 SIPP, 1.5 million children experienced their parent’s divorce in 2000, a rate of 21 per 1,000 children (Kreider & Fields, 01-Stewart.qxd 11/28/2006 6:29 PM Page 17

Brave New Stepfamilies 17

2005). The majority of divorced people remarry, and remarriage rates are higher among men than women (Kreider & Fields, 2002): Projections based on the 1996 SIPP indicate that 88% of men and 66% of women will remarry after divorce. According to the 2001 SIPP, in 25% of currently married couples, one or both spouses have been previously married, 13% of men and 14% of women report having been married twice, and 3% of each reported having been married three of more times (Kreider, 2005). According to the 2001 SIPP, 57% of women and 17% of men who divorced in the previous year were living with children under the age of 18 (Kreider, 2005). Our most recent estimate, based on the 1980 CPS, tells us that roughly two thirds of remarriages involve children from a previous marriage and the majority of these (75%) have stepchildren living in the household (Cherlin & McCarthy, 1985). Thus a traditional stepfamily is born. More recent data from the 2001 SIPP indicate 15% of children under age 18 are living in a married stepfamily and that about one in 10 households containing children are stepfamilies (Kreider & Fields, 2005). In three quarters of divorce cases involving children, the mother receives sole physical custody of their children (Cancian & Meyer, 1998). The 2001 SIPP indicates that 80% of children who experienced their parent’s divorce in the previous year live with their mothers (Kreider & Fields, 2005). Resident stepchildren are therefore most often the biological children of the wife, so that the majority (between 80 and 90%) of traditional stepfamilies contain a stepfather rather than stepmother (Moorman & Hernandez, 1989). A traditional definition of a stepfamily would indicate that the vast majority of stepparents are men. According to the 2001 SIPP, one in 10 resident fathers has stepchildren, compared with 2% of resident mothers (Kreider & Fields, 2005).

Stepfamilies Created by Nonmarital Childbearing

Current estimates of stepfamilies often include first marriages in addition to remarriages. However, available sources do not distinguish stepfamilies that are remarriages from stepfamilies that are first marriages. Using data from the June 1990 CPS of the U.S. Census (newer estimates are not available), Norton and Miller (1992) estimated that 5.3 million married-couple family households (a first marriage or remarriage) contained at least one stepchild under age 18. This number represents 21% of all married-couple households with children, an increase from 16% in 1980 (Norton & Miller, 1992). 01-Stewart.qxd 11/28/2006 6:29 PM Page 18

18 BRAVE NEW STEPFAMILIES

Norton and Miller (1992) also examined the prevalence of stepfamily living from the vantage point of children. Of children under 18 years of age living in a married-couple household, almost one in five children (19%) were not living with a biological mother and biological father. About 15% were living with a biological mother and stepfather, and 1% of children were living with a biological father and stepmother. Hofferth (1985) analyzed data from the Panel Study of Income Dynamics (PSID) and found that the proportion of children living with a stepparent (within marriage) has increased steadily in the second half of the twentieth century. She estimated that 25% of White and 27% of African American children born in 1980 would live with a stepparent by age 18 compared with 3% and 7% born in 1950–1954. With rates of divorce and nonmarital childbearing remaining relatively stable, and marriage and remarriage rates declining somewhat (the difference being made up by increasing cohabitation), it is reasonable to assume that the proportion of married couples with stepchildren would be only slightly lower today than a decade ago (Bianchi & Casper, 2000).

Cohabiting Stepfamilies

Limiting stepfamilies to marriage underestimates the prevalence of stepfamilies, because couples with children are increasingly likely to cohabit. For example, 40% of births between 1990 and 1994 that occurred to “single mothers” were actually born to women cohabiting with the child’s father (Bumpass & Lu, 2000). Children residing with cohabiting couples are often the product of a previous romantic relationship, however. According to the 2001 SIPP, 11.4% of children under age 18 living with a “single” biological (or adoptive) parent also resided with the parent’s cohabiting partner: 9.4% lived with a mother and her partner and 2% lived with a father and his partner (Kreider & Fields, 2005). If “unmarried partners” were to be considered cohabiting stepparents, according to these data, over one in four children (27.5%) living with a stepparent would be living with a stepparent who is not married to the biological parent. This figure is similar to a previous estimate based on the 1987–1988 NSFH, which indicated that in 25% of stepfamilies the parents are not legally married (Bumpass et al., 1995). Because two thirds of children enter stepfamilies through cohabitation rather than marriage, not accounting for cohabitation greatly underestimates the duration (the length of time the family has been together) of stepfamilies as well (Bumpass et al., 1995). 01-Stewart.qxd 11/28/2006 6:29 PM Page 19

Brave New Stepfamilies 19

Census 2000 also provides estimates of stepchildren and households with stepchildren based on whether respondents reported a “stepchild.” This is an advance, because people who reported a stepchild could include those who were married, never married, or formerly married, reflecting the more recent usage of the term (Kreider, 2003). Yet an important limitation is that the census only measures the number of stepchildren of the householder (the person who filled out the survey) and cannot, therefore, capture stepchildren of other people in the household. Because of the way the data were collected, these estimates probably capture only about two thirds of all the stepchildren living with a stepparent (Kreider, 2003). For instance, Kreider and Fields (2005) used the SIPP and estimated that, in 2001, 5 million children under 18 (one in 10 children) were living with at least one stepparent, which is somewhat higher than the census figure. Another problem is that the start and end dates of cohabitation are often undefined and cohabitation is often short in duration. Measuring cohabitation is in general difficult, and large social surveys probably underestimate it (Manning & Smock, 2005).

Multihousehold Stepfamilies

The estimates discussed above are restricted to stepfamilies in which all the members reside in the same household. Only relatively recently have national data sources collected information on members of stepfamilies who live in different households. For instance, many children who live in single- mother households visit their father and stepmother on the weekends. Although nonresident parents are often thought of as “deadbeat dads” and “deadbeat moms,” there is actually a lot of diversity in nonresident parents’ involvement with biological children in other households. Data from the 1997 National Survey of America’s Families (NSAF) indicate that, whereas one third of children with a nonresident parent have no contact with him or her, one third see their nonresident parent once a week or more (Stewart, 2004). Because most nonresident parents eventually cohabit and/or marry, these visits can include stepparents. However, few studies allow researchers to assess nonresident stepparent-stepchild relationships directly. A survey that allows the identification of nonresident stepparents and stepchildren is the 1987–1988 NSFH, which indicates that in almost one quarter of married and cohabiting stepfamilies with resident stepchildren, at least one of the parents has nonresident biological children (Thomson, 1994). 01-Stewart.qxd 11/28/2006 6:29 PM Page 20

20 BRAVE NEW STEPFAMILIES

Wineberg and McCarthy’s (1998) analysis of these data reveals that 28% of remarried couples and 45% of previously married cohabiting couples have both resident and nonresident stepchildren. Stewart (2001) focuses on stepparents and finds that roughly half (47%) of married and cohabiting stepparents have stepchildren living in other households and an additional 6% have both resident and nonresident stepchildren. This means that less than half (48%) of stepparents live with their stepchildren full time. Broadening the definition of a stepfamily to include nonresident stepchildren and stepparents would mean a near doubling of the stepfamily population! As is the case for cohabitation, the measurement of residential and nonresident stepfamily members is problematic because stepchildren’s living arrangements can be unstable (Maccoby & Mnookin, 1992).

African American Stepfamilies

Few studies have focused on racial and ethnic diversity in stepfamilies. One explanation is that racial and ethnic minorities tend to be underrepre- sented in traditional stepfamilies based on divorce and remarriage. Thomson (1994) used the 1987–1988 NSFH to show that stepfamilies, when defined broadly to include first marriages and cohabitation, are disproportionately composed of racial and ethnic minorities when compared with original, two- parent families (20% vs. 13%). Newer studies that also take a broad view of stepfamilies indicate that stepchildren are disproportionately African American (Kreider, 2003; Kreider & Fields, 2005). The 2001 SIPP indicates that 15% of Black children living with two parents live with a married or cohabiting stepparent, compared with 9% of White children (Kreider & Fields, 2005). Stepgrandparenthood is also much more common among African Americans than among Whites or Hispanics (Szinovacz, 1998a). Estimates commonly distinguish Black and White children, but little is known about the extent of stepfamily living among other racial and ethnic groups. For example, I am aware of no estimates of Latino, Asian, or American Indian stepfamilies.

Stepfamilies With Gay or Lesbian Parents

Stepfamilies with gay or lesbian parents are also absent from most estimates of stepfamilies. In general, gay and lesbian parents tend to be hidden because many do not disclose their sexual orientation, even to their children 01-Stewart.qxd 11/28/2006 6:29 PM Page 21

Brave New Stepfamilies 21

(Lynch, 2000). This makes the measurement of this population a challenge. Nevertheless, it has been estimated that one out of nine cohabiting couples is a same-sex couple (Simmons & O’Connell, 2003). The 1990 census indicates that 17% of gays and 29% of lesbians have been in a heterosexual marriage (Black, Gates, Sanders, & Taylor, 2000). The 2000 Census indicates that 33% of female same-sex households and 22% of male same-sex households (roughly 165,000) contain children from previous relationship or the current union (Simmons & O’Connell, 2003). Unfortunately, that study does not distinguish children from previous and current relationships. Whereas gay stepfamilies are mostly noncustodial, lesbian stepfamilies usually involve resident children (Berger, 2000).

Stepfamilies With Adult Stepchildren

The vast majority of estimates focus on stepfamilies with children who are less than age 18. From the perspective of children’s well-being, a heavier focus on younger children is warranted. However, stepfamily relationships do not disappear when children reach their eighteenth birthday. Demographer Paul Glick (1989) was the first to project children’s lifetime chances of a living in a stepfamily through adulthood (including children who did not live with their stepparents). Using data from the 1987 CPS, he estimated that under this wider definition of stepfamilies, over 50% of children in the United States would have at least one stepparent in their lifetime. This compares with Bumpass et al.’s (1995) analysis of the 1987–1988 NSFH, which put this figure at 30% through age 18. Results from Census 2000 show that there are 1.1 million children 18 years of age and older living with a stepparent, which represents 6% of all adult children (Fields, 2003). It has become vital to consider cohabitation when estimating stepfamilies with adult stepchildren. Cohabitation may even be replacing remarriage for older couples in Europe (De Jong Gierveld & Peeters, 2003). At the time of the last U.S. Census, there were over one million adults age 50 and older who were cohabiting (Brown, Bulanda, & Lee, 2005). In recent decades, cohabitation grew at a much faster rate for older than younger persons. Census data indicate the rate of cohabitation for those 60 and over tripled during the 1980s but only doubled for those under the age of 40 (Chevan, 1996) Because most men and women of that age have children (some may even reside with them), these cohabiting unions would form a new kind of stepfamily. 01-Stewart.qxd 11/28/2006 6:29 PM Page 22

22 BRAVE NEW STEPFAMILIES

Other researchers have looked at the prevalence of stepsiblings among adult men and women. Findings based on the 1987–1988 NSFH indicate that 4% of adults with siblings have stepsiblings and no biological siblings, and 9% have both full siblings and stepsiblings (White, 1994). This percentage is probably greater among more recent cohorts of adults, who were more likely to experience their parents’ divorce. Another perspective is to look at the number of stepgrandparents. Stepgrandparents may result from either the grandparent or the grandparent’s child getting a divorce and repartnering. About two fifths (39%) of married respondents with adult stepchildren have a stepgrandparent in the family, and this figure would be more if the researcher had also included cohabiting stepfamilies (Szinovacz, 1998a).

Former Stepfamily Members

Most estimates of stepfamilies don’t include people who used to be stepparents or used to be stepchildren and who no longer are because their union, or their parents’ union, has dissolved. This scenario is not infrequent given the instability of remarriage and cohabitation (Bumpass & Lu, 2000; White & Booth, 1985). Remarried couples, for instance, have a higher risk of divorce than first married couples (Goldstein, 1999). However, stepparent-stepchild relationships do not necessarily disappear simply because the stepparent and biological parent’s union has dissolved. For example, the 2001 SIPP indicates that thou- sands of children reside with just a stepparent and no biological parent (Kreider & Fields, 2005). Estimates also generally don’t include stepchildren who have been legally adopted by their stepparents (for exceptions see Moorman & Hernandez, 1989; Norton & Miller, 1992), which is common especially among stepfamilies created by a nonmarital birth (Mahoney, 1994b). Finally, a child may be a member of a stepfamily but not be a stepchild himself or herself. These are children whose family is technically “intact” (i.e., they have both biological parents in the home) but who have half-siblings born of their parents’ previous relationships.

SUMMARY

What can we conclude from the above discussion? First, most figures pertain- ing to stepfamilies are gross underestimates of the extent to which stepfamily living has permeated society. Psychologists Bray and Kelly (1998) suggest that 01-Stewart.qxd 11/28/2006 6:29 PM Page 23

Brave New Stepfamilies 23

stepfamilies will outnumber first-married families by the year 2007. In an article for American Demographics magazine, demographer Paul Glick states that, “One out of every three Americans is now a stepparent, a stepchild, a stepsibling, or some other member of a stepfamily,” and that “more than half of Americans today have been, are now, or will eventually be in one or more step situations during their lives” (qtd. in Larson, 1992, p. 36). Berger (1995) notes that “stepfamilies have multiplied to the point that some writers refer to them as the future traditional American family” (p. 35). While these claims may be somewhat exaggerated, it is probably the case that the majority of Americans are connected to or will be connected with a stepfamily in some way. However, it is probably not going to be in the way that most of us imag- ine. The stepfamilies of today, and the stepfamilies of the future, are not the stepfamilies of the past. The goal of this book is to describe why this transfor- mation is occurring and discuss what these changes mean for stepfamily life in the 21st century.

NOTE

1. The role of adoption with respect to stepfamily life is ambiguous. Families in which both partners legally adopt a nonbiological child (e.g., through an adoption agency) are not considered stepfamilies. However, partners sometimes legally adopt the biological (or adopted) children of their spouse. Stepfamily scholars are not in agreement as to whether this situation constitutes a stepfamily. Some researchers classify adopted stepchildren as “stepchildren” (Moorman & Hernandez, 1989; Norton & Miller, 1992). Other researchers consider these children the shared children of the couple because after the adoption stepparents become legally responsible for their stepchild (Bray & Berger, 1993; Ganong & Coleman, 2004). Legal adoption may not be an option for stepfamilies with gay or lesbian parents because in some states, children cannot legally have “two mothers” or “two fathers” (Fine, 1994). Stepparent adoption is covered in Chapter 4.